Adaptive Route Control for a Vehicle using Fuzzy Logic and Genetic Algorithms

D. Cruz and S. Cardona (Columbia)


Fuzzy Logic, Genetic Algorithms, Microprocessor-based card, Fuzzy Rule Sets, Adaptive Control.


The purpose of this paper is to present the development of a route control system for a vehicle that adjusts itself to mechanical variations or changes in its environment. This control was developed using Fuzzy Logic and the system adaptability relies on Genetic Algorithms. With this objective in mind, a microprocessor-based card was implemented; this card holds the fuzzy logic route control system assigned to follow a predetermined path, and receives data from the sensors placed on the mobile robot via a radio link. Genetic algorithms are used for the on-line modification of the fuzzy rule set in the route control system. These algorithms find the very best rule set for the fuzzy route control system given some determined conditions for the environment on which the car is, or based on the model of the vehicle. When the path following is taking place, the system optimizer picks up information from the route control system, that by means of a genetic algorithm allows to build a model for the vehicle which represents its physical and environmental conditions. When this model is done, a second genetic algorithm finds the rule set that suits the newly found model so the fuzzy route control can perform in an optimal manner.

Important Links:

Go Back